Context This kernel contains data contains features generated in my kernel: https://www.kaggle.com/artgor/even-more-features using data from this competition: https://www.kaggle.com/c/LANL-Earthquake-Prediction I have commented some of the code in the kernel because otherwise it would run for too long and hit the time limit. So this dataset contains all the 987 features for train and test data from the kernel and the train target variable. **UPD**: Added 979 more features in separate files. I used this great kernel https://www.kaggle.com/tarunpaparaju/lanl-earthquake-prediction-signal-denoising and denoised data before calculating statistics.